Unjustified Enrichment: Key Issues in Comparative Part 4 pot

The importantpoint for present purposes is that in these systems fraud and duress couldaffect the validity of all declarations of intent and legal transactions oracts,36 by rendering the

any otherartifice’.11As far as South African law is concerned, the meaning

of ‘fraud’ has long been uncertain, due to a strange convergence of and common-law influences At first it was equated with the very broad

civil-civilian concept of dolus malus, which covered all sorts of cases involving

intentional deception, cheating orcircumvention.12 However, under theinfluence of the common law, ‘fraud’ later came to be interpreted morerestrictively, along lines very similar to that of the English law as set outabove Thus, in all these systems, fraud, or its equivalent, at least consists

in knowingly or intentionally making a false representation Although itcould still covergross negligence undercommon law, it is clearthat itcertainly does not cover mere negligent or innocent misrepresentation

It is as a descriptor of the basic ‘fact pattern’ of particularly serious types

of misrepresentation that the concept of ‘fraud’ will be used here.13 Thedifferences in where the cut-off point lies can for present purposes beregarded as of secondary importance.14

However, a difference which is especially important in the context of thelaw of unjustified enrichment is the way in which these different systemsperceive the relationship between fraud and certain other improper ways

of obtaining consent In English contract-law texts, fraud is viewed as aspecies of misrepresentation,15and therefore as a specific means of induc-ing an error It is also not traditionally grouped together with duress Thisstands in contrast to the position in German and Dutch law, which reflect

11 Art 3: 44(3) BW; also see A S Hartkamp, Mr C Asser’s Handleiding tot de Beoefening van Het Nederlands Burgerlijk Recht – Verbintenissenrecht (1997), vol II, nn 199–204, where it

is indicated that the expression ‘through any other artifice’ (door een andere kunstgreep) should be interpreted restrictively Bedrog essentially requires an intention to

deceive.

12 See Lubbe, ‘Voidable Contracts’, 265 referring to the definition of dolus malus in Ulp.

D 4, 3, 1, 2.

13 See generally Zweigert and K ¨otz, Introduction, 425 ff In some systems, ‘specially

deceptive practices’ are required, but as Zweigert and K ¨otz point out, ‘[e]ven so, it is

admitted on all hands that a simple lie can constitute deceit’ (ibid., 425).

14 In the English common law and in South African law, these differences were particularly important in determining whether a claim for damages would be available: traditionally, such a claim could only succeed in those cases of

misrepresentation which amounted to fraud; rescission, on the other hand, could be obtained even in the event of innocent misrepresetation; see Lubbe, ‘Voidable Contracts’, 270 ff.

15 See generally Treitel, Law of Contract, 317 ff.; Anthony Guest (ed.), Chitty on Contracts

(28th edn, 1999), vol I§§ 6-001 ff., 6-045 The law of misrepresentation covers all cases

of error induced by misrepresentation In fact, it has been said that the term

‘mistake’ could be used to refer only to those cases of error not caused by

misrepresentation; see Zweigert and K ¨otz, Introduction, 421.

the traditional civilian preference for linking these notions.16 There, nocent and negligent misrepresentation are generally dealt with in thecontext of the law relating to error, whereas the more serious cases of

in-fraudulent representation (the terrain of the law of arglistige T¨auschung and bedrog) are grouped with duress.17 But why view fraud in this light?

If it in any event causes an error, and if error is a ground for relief, whynot simply deal with it as a special case of error? Apart from obvioushistorical considerations, it would seem that the answer lies in the fea-ture of fraud identified earlier, namely that the means used to induceerror in these cases are so seriously improper that the normal rules ap-plying to the restitution of transfers made in error are inappropriate It

is recognised that the victim deserves special protection This could beprovided, for example, by not requiring that his error be material or fun-damental In other words, he could be allowed to escape from liabilityeven though the fraud simply caused an error in motive.18 It also meansthat he should be provided with a fuller spectrum of remedies, and that

he should be able to recover more than the person who merely acted der an innocent or negligent misrepresentation But these are issues morerelevant to the effect of fraud than its content, and will be returned tolateron

un-2 Duress or unlawful threats

The (originally Norman French) term ‘duress’ is derived from the commonlaw Traditionally, it only covered the situation where a person was un-lawfully subjected to actual or threatened personal physical harm.19Thismeans that in some other cases where a person was subjected to pressure,relief had to be provided on different grounds – usually ‘undue influence’

in equity.20 However, the scope of duress has broadened over the years

so as also to accommodate harm to economic interests, and nowadaysthe difference between the two concepts is rather indistinct German andDutch law, on the otherhand, have avoided these problems Although a

16On the historical background to the law of dolus and metus see Zimmermann, Law of Obligations, 664 ff.

17 On the relationship between error and misrepresentation in English law, as opposed

to German and Dutch law, see Michael H Whincup, Contract Law and Practice: The English System and Continental Comparisons (3rd edn, 1996), §§ 11.49, 11.56 ff.

18Zweigert and K ¨otz, Introduction, 425; Englard, ‘Restitution of Benefits’, § 5-81 See further Larenz, Allgemeiner Teil, § 20; Hartkamp, Mr C Asser’s Handleiding, n 199.

19See Beatson, Anson’s Law of Contract, 271–2; Treitel, Law of Contract, 375; Skeate v Beale (1841) 11 A & E 983; Cumming v Ince (1847) 11 QB 112 at 120; Biffen v Bignell (1862) 7 H

& N 877.

20See Treitel, Law of Contract, 375, 378; Zweigert and K ¨otz, Introduction, 428.

traditionally narrow conception of metus required that at times additional

relief had to be provided on different grounds,21it is recognised nowadaysthat duress covers all cases where a person is unlawfully subjected to ac-tual or threatened harm.22Because the fearhas to emanate from the per-son who made the threat, there can be no duress if the pressure is (merely)the consequence of some situation of need ordistress.23 No distinction isdrawn between forms of duress according to the nature of the interests(forexample, physical oreconomic) harmed.24The key question is simplywhether the threat is unlawful South African law again bears the im-print of both civil- and common-law influences: it received the civil law of

metus (translated as ‘duress’), which was flexible enough to accommodate

a broad range of threats of harm, but under the influence of the commonlaw it somehow adopted a restrictive approach to providing relief in whatcould be termed cases of ‘economic duress’ It further felt the need to

supplement the law of metus with the common law of undue influence.

Curiously though, the justification for this development was not sought in

the civil law of metus (and especially ‘reverential fear’ or metus reverentialis).

In fact, it was with reference to the idea that in civil law restitutio in grum would be provided in cases of dolus (interpreted in the broad sense

inte-indicated above) that the reception of undue influence from the commonlaw was facilitated.25

As in the case of ‘fraud’, it is not necessary to analyse the meaning of

‘duress’ in great detail However, one important observation needs to bemade It deals with the much-disputed basis forproviding relief.26 Is thevictim protected because his mental ability to make a decision is affected

by the duress, or is it because of some other ground, such as an unlawfullimitation of the freedom of choice through subjecting a person to threats

of harm? Civilians as early as Paulus have held that the basis for relief isnot a defect in mental ability: what is willed undercompulsion, none

21 Most notably the law of the condictiones (see Jacques E du Plessis, Compulsion and Restitution (unpublished Ph.D thesis, University of Aberdeen, 1998), 16 ff., 40 ff., 60 ff.,

121 ff., 134 ff.; John P Dawson, ‘Economic Duress and the Fair Exchange in French

and German Law’, (1937) 11 Tulane LR 345, 348.

22 See Zweigert and K ¨otz, Introduction, 428 On Drohung in German law see Kramer in: M¨unchener Kommentar, § 123, n 33; and in Dutch law art 3:44(2) BW.

23 See Zweigert and K ¨otz, Introduction, 428; Kramer in: M¨unchener Kommentar, § 123, n 33.

In cases of vis absoluta, where the person is being physically overpowered, he obviously does not act at all See Zweigert and K ¨otz, Introduction, 428; Kramer in: M¨unchener Kommentar, § 123, n 32.

24 On the historical background see Zimmermann, Law of Obligations, 659.

25 See Lubbe, ‘Voidable Contracts’, 286 ff.

26 For a historical overview, see Gordley, ‘Contract in Pre-commercial Societies’,§§ 2–4 ff.

the less is willed (coactus volui, tamen volui).27 The victim could be quiterational, and not even be afraid: his problem is that he cannot freely decidewhat he wants to do because the available choices have been restrictedunlawfully.28The emphasis is then on the nature of the methods used toinfluence his will, and not on a deficiency in the will itself In the commonlaw, by contrast, it has been said that duress ‘overbears’ the victim’s will,29thereby influencing his mental ability to make a decision However, somecommon lawyers have criticised this approach; after all, if all that has to

be proven is an ‘overborne will’, even the victim of lawful pressure should

be able to escape liability on grounds of duress.30

From the overview above it should be apparent that fraud and duressdeal with serious violations of individual autonomy In the case of fraud,the victim’s freedom of choice cannot be exercised properly, because hewas made to act on wrong information ‘conveyed’ in a particularly un-acceptable manner, while in the case of duress his freedom of choice isrestricted by actual or threatened harm It then stands to reason thatthese violations justify a strong measure of protection – stronger at leastthan the case where the actions are prompted by a spontaneous error

or the sorts of pressures which are part of daily life It will now be ered how the law of restitution or unjustified enrichment can fulfil thisfunction

consid-III The effect of fraud and duress

In the course of the eighteenth century, German legal scholars devised the

concepts of the ‘declaration of intent’ (Willenserkl¨arung) and the ‘legal action’ (Rechtsgesch¨aft).31In essence, a declaration of intent is a declaration

trans-27Paul D 4, 2, 21, 5 See Zimmermann, Law of Obligations, 652 ff.; Du Plessis, Compulsion and Restitution, 6 ff.

28See Du Plessis, Compulsion and Restitution, 161 ff.

29See Pao On v Lau Yiu Long [1980] AC 614.

30See Lynch v DPP of Northern Ireland [1975] AC 695 B–C; Patrick S Atiyah, ‘Economic Duress and the “Overborne Will”’, (1982) 98 LQR 197; Nicholas Seddon, ‘Compulsion in Commercial Dealings’, in: P D Finn, Essays on Restitution (1990), 138, 142 ff.; Beatson, Anson’s Law, 273 ff.

31See Werner Flume, Allgemeiner Teil des B¨urgerlichen Rechts (3rd edn, 1979), vol II, §§ 2, 4 Flume indicates that the Roman concepts such as actus and negotium could not fulfil

such a function The notions were never received in South African law, whose civilian roots lie in seventeenth- and eighteenth-century Roman-Dutch law Modern

introductory textbooks on South African law refer to concepts such as ‘legal act’ or

‘juristic act’ (regshandeling), but only as pedagogical tools; see, forexample, H R Hahlo and E Kahn, The South African Legal System and its Background (1968), 100 ff.

indicating that certain legal consequences are intended32 and it is tial forthe validity of many legal transactions, of which the contract isbut one.33These concepts (which were unknown in Roman law, and whichare still of scant relevance in modern South African law) have influencedothercivil-law codes Dutch law, forexample, recognises similarconcepts,

essen-most notably that of the ‘legal act’ (rechtshandeling).34 These are acts thatproduce a legal effect or consequence because they were intended to do

so.35This intention has to be manifested in a declaration The importantpoint for present purposes is that in these systems fraud and duress couldaffect the validity of all declarations of intent and legal transactions oracts,36 by rendering them voidable.37 This means that the person enti-tled to annul can decide whetherto uphold the validity of the legal act.The effect of annulment is retrospective, so that the legal relationship ofthe parties is restored to the state in which it was before the act wasperformed.38

But let us move from the general to the particular Of all the types ofdeclarations of intent or acts that can be influenced by fraud and duress,the present concern is with those relevant when determining liabilitybased on unjustified enrichment For present purposes, the most impor-tant of these are declarations or acts aimed at (1) creating contractualobligations, (2) achieving fulfilment orperformance of all types of obliga-tions and (3) transferring ownership As far as the difference between the

32 See, generally, Larenz, Allgemeiner Teil, § 19, and the authorities quoted there.

33 Cf Zweigert and K ¨otz, Introduction, 348 ff.

34 See A R Bloembergen, Rechtshandeling en Overeenkomst (1995), 2 ff on the notion of a rechtshandeling and Flume, Allgemeiner Teil, §§ 2, 4 on the influence of the German Civil

Code on Meijers’s draft of the New Dutch Civil Code, and also on its influence on the position in France, Italy and Greece.

35 See art 3:33 BW and generally Hartkamp, Mr C Asser’s Handleiding, n 2.

36 § 123(1) BGB The reason why a declaration of will, and not only a juristic act, is

affected is that the compilers of the BGB wanted to provide maximum protection to

a person acting under duress (Benno Mugdan, ‘Denkschrift zum BGB’, in: Benno

Mugdan (ed.), Die gesammten Materialien zum B¨urgerlichen Gesetzbuch f¨ur das deutsche Reich

(1899), vol I, 834).

37 See§ 123(1) BGB; art 3:44(1) BW; Arthur S Hartkamp and Marianne M M Tillema, Contract Law in the Netherlands (1995), §§ 36, 99 On the historical background

regarding the consequences of voidability as opposed to voidness, see Zimmermann,

Law of Obligations, 660, 671 ff It is only in certain cases of fraud that the error it

causes is so material that the act should be regarded as void (see section II, 1, above,

on the relationship between misrepresentation and error) In the case of absolute

duress or vis absoluta (e.g where a person’s hand is held and thus forced to make a

signature) there obviously is no declaration of intent at all This renders any act affected by it automatically void.

38 See§ 142(1) BGB; art 3:53 BW.

first two categories is concerned, attention is usually focused on the ation where a contract is concluded and restitution is sought of what wasperformed thereunder This is, of course, an important category of cases,but it should not be ignored that fraud and duress can also influence theconferring of a benefit outside a contractual context For example, they

situ-may move a person to perform ex lege obligations, oreven to engage in acts

that do not involve performance at all, such as to make a bequest, declare

a dividend or grant a licence These cases do not concern the validity of acontract between the victim and the wrongdoer, but rather some other actwhose validity is suspect As faras the difference between the second andthird categories is concerned, it is important from a civilian perspective

to distinguish between the intention to fulfil an obligation (that, is to form), and the intention to transfer ownership The position in practicenormally would be that the person who intends to transfer ownership ofsomething does so in order to fulfil an obligation, but this need not bethe case – he might, forexample, intend to make the transfernow with aview towards creating a loan agreement or donation in the future A ques-tion that is of particular importance in this regard is the extent to whichfraud or duress may prevent ownership from passing in the first place,and so entitle the owner to vindicatory relief, rather than relief based onunjustified enrichment The last category therefore deals with the border-line between the laws of property and unjustified enrichment But firstconsider the borderline between the laws of unjustified enrichment andcontract

per-1 The effect of fraud and duress on the validity of contracts

(a) Rescission and restitutio in integrum

In the civilian systems under review, the question whether fraud andduress affect the validity of a contract has not traditionally been the con-cern of the law of unjustified enrichment From a historical perspective,this is perfectly understandable In Roman times, the praetor did not try to

counteract fraud or duress through developing the law of the condictiones.39

He dealt with the problem head-on by exercising his extraordinary powers

to order restitutio in integrum, thus ensuring that both parties were restored

to their previous position, and by awarding an actio quod metus causa, which

apparently was aimed at inducing the victim to make restitution through

39A possible, and problematic, exception is the extorted stipulatio; see Pomp D 12, 5, 7;

Du Plessis, Compulsion and Restitution, 23 ff.

subjecting him to fourfold damages if he did not.40 The fact that

restor-ing parties in integrum generally took place in the context of rescission

of contracts explains how in some systems restitutio in integrum came to

be regarded as ‘contractual’ in nature, something tailor-made for cases ofimproperly obtained consent This is still the position in modern SouthAfrican law, which (mainly due to differences in the measure of recovery)expressly rejects the idea that a duty of restitution in cases of fraud orduress is based on unjustified enrichment.41

However, other modern civilian systems view matters rather differently.Already at the end of the nineteenth century, the compilers of the German

code regarded it as unnecessary to recognise a remedy styled restitutio in integrum.42 Where a declaration of intent was rescinded on grounds offraud or duress, any juristic act of which this declaration of intent formedpart would be invalid If such a juristic act was supposed to act as the le-

gal ground or causa of a transfer, that legal ground would be regarded as

neverhaving existed.43 It is only here that the law of unjustified ment would enterthe scene.44By determining that the contract is invalid,the law regarding the validity of declarations of intent and legal transac-tions has already done the hard work of indicating that a performance

enrich-is retained without legal ground The law of unjustified enrichment only

40 See Berthold Kupisch, In integrum restitutio und vindicatio utilis bei

Eigentums¨ubertragungen im klassischen r¨omischen Recht (1974); Zimmermann, Law of Obligations, 656 ff.; Du Plessis, Compulsion and Restitution, 13 ff.

41 Foran exposition of these views and criticism, see D P Visser, ‘Rethinking

Unjustified Enrichment: A Perspective of the Competition between Contractual and

Enrichment Remedies’, [1992] Acta Juridica 203, 211.

42 The reasons for not incorporating a provision dealing with restitutio in integrum

mainly relate to problems with its application in earlier times, and changed

procedural views and institutions (Benno Mugdan, ‘Motive’, in: Mugdan, Die

gesammten Materialien, vol II, 566 ff.) It was considered unnecessary because the

provisions on unjustified enrichment could perform this role satisfactorily (see

Visser, ‘Rethinking Unjustified Enrichment’, 215 ff.) The actio quod metus causa

likewise was dropped because general provisions were adopted on delict and unjustified enrichment, and there was a desire not to burden the code with

unnecessary provisions (Mugdan, ‘Motive’, 423) The same was true of the condictio ex iniusta causa, which was used in the German ius commune law to reclaim illegally extorted bestowments (see Detlef K ¨onig, Ungerechtfertigte Bereicherung: Tatbest¨ande und Ordnungsprobleme in rechtsvergleichender Sicht (1985), 47).

43 On the retrospective operation of rescission, and exceptions thereto, see Theo

Mayer-Maly, in: M¨unchener Kommentar zum B¨urgerlichen Gesetzbuch (3rd edn, 1993), vol I,

§ 142, n 14 ff.

44 See Manfred Lieb, in: M¨unchener Kommentar zum B¨urgerlichen Gesetzbuch (3rd edn, 1997),

vol V,§ 812, nn 144, 148; Zimmermann, ‘Unjustified Enrichment’, 407; Englard,

‘Restitution of Benefits’,§ 5-9.

has to regulate what has to be restored in so far as this has not alreadybeen done by specific rules of the law of contract However, this still leavesthe problem of ensuring some reciprocity in the restitution of what theparties performed One solution is simply to say that rescission would

not be granted if restitutio in integrum is impossible.45 Such an approachcould operate rather harshly on the victim, and therefore cannot be fol-lowed without exception In this regard, it is of interest that Dutch lawrecognises the qualification that the victim may be protected by allowing

‘equitable monetary adjustments’.46

However, in modern German law, the situation is viewed somewhat ferently In essence, the inability to provide restitution does not precluderescission; the victim can obtain rescission, but the entitlement to andmeasure of enrichment-based relief is affected in three ways First, under

dif-§ 814 BGB the victim is not entitled to relief, if at the time of performance

of a contract he knew that he was not obliged to perform Thus, if he knewthat the validity of the contract was tainted by fraud, but still performed,his enrichment claim would be barred It can justifiably be asked why hedid not refuse to perform when he had the chance (It is obvious that thevictim of duress has to be treated differently: it is precisely because of thecompulsion that he cannot refuse even if he knows that he is not obliged

to perform Duress indicates that an enrichment remedy should not bebarred.) In this regard it has been argued that the reason for excluding aclaim where the transferor knew that the transfer was not due is to pre-

vent him from acting contrary to his previous conduct (venire contra factum proprium).47Secondly, under§ 819(1) BGB, the wrongdoer is not entitled to

plead loss of enrichment or change of position if he was aware that he wasnot entitled to keep the enrichment In other words, he cannot plead thathis liability should be restricted to what remains in his hands, instead

45See, forexample, in the context of Scots law, Du Plessis, Compulsion and Restitution,

92 ff., 167 ff.

46 Art 3:53 BW On the position in the common law, see the contribution of Mindy

Chen-Wishart to this volume Although restitutio in integrum is regarded as a condition

to rescission (see Erlanger v New Sombrero Phosphate Co (1878) 3 App Cas 1218 at 1278),

it is also acknowledged that the innocent party’s right to rescind is not automatically barred due to his own inability to provide restitution (Robert Goff and Gareth Jones,

The Law of Restitution (5th edn, 1998), 273 ff.; Treitel, Law of Contract, 350-1; Englard,

‘Restitution of Benefits’,§§ 5-74, 5-88).

47See PeterSchlechtriem, in: OthmarJauernig (ed.), B¨urgerliches Gesetzbuch (7th edn,

1994),§ 814, nn 1, 5; see further Markesinis, Lorenz and Dannemann, Law of Contracts, 736 On the influence of duress on the recoverability of undue transfers

outside the contractual context, see section III, 2, (a) and (b), below.

of the full extent of what he received Thirdly, the wrongdoer is not

pro-tected by the Saldotheorie, which seeks to ensure reciprocity in the return

of transfers.48 Although the BGB seems to have been drafted in a waywhich indicates that one is dealing with separate claims (this is the way

in which the old ‘doctrine of the two enrichment claims’ or tionenlehre also understood it), the Saldotheorie ‘combines the legal fates’

Zweikondik-of these claims through a process Zweikondik-of adjustment.49 In essence, whenever

a person’s performance has been destroyed while being retained by his

contracting partner, he can raise the Saldotheorie against the partner’s

en-richment claim The defendant/enen-richment debtor can only counterclaim

to the extent that he is able to provide restitution If he lost somethingworth more than his counterclaim, he in effect has no claim However,since the wrongdoer who defrauds another is not entitled to the protec-

tion of the Saldotheorie, the victim should be entitled to reclaim his full

performance.50

(b) The nature of enrichment-based remedies

(i) Leistungskondiktion and its scope of application

Against the background of the rescissory actions and restitutio in integrum,

it can now be investigated what type of enrichment-based remedy willprovide the victim with relief In German law the code itself is ratherunhelpful All§ 812(1), first sentence BGB states is that a person who ob-

tains something without legal ground at the expense of another through atransfer by that person or in any other way is liable to return it However,

it will be noticed that in the code a distinction is drawn between obtaining

something through a transfer (Leistung), and obtaining it in any otherway.

Nowadays, this distinction is regarded as crucial.51A typology has been

de-veloped that places all enrichment claims involving a Leistung ortransfer

48 See Englard, ‘Restitution of Benefits’,§§ 5-90, 5-278 ff.; DieterMedicus, Schuldrecht II: Besonderer Teil (9th edn, 1999), § 129; Zimmermann and Du Plessis, ‘Basic Features’,

40 ff.

49 See Medicus, Schuldrecht II, § 129.

50 See BGHZ 53, 144; Englard, ‘Restitution of Benefits’,§ 5-279 However, some

adjustment based on contributory fault is possible For criticism of the view that

fraud should be an exception to the Saldotheorie in cases where there is no causal link

between the misrepresentation and the harm giving rise to the law of enrichment, see Markesinis, Lorenz and Dannemann, 765.

51 See the pioneering works of Walter Wilburg, Die Lehre von der ungerechtfertigten Bereicherung nach ¨osterreichischen und deutschem Recht (1934), and Ernst von Caemmerer,

‘Bereicherung und unerlaubte Handlung’, in: Festschrift f¨ur Ernst Rabel (1954), vol I,

333 ff.; Markesinis, Lorenz and Dannemann, 717 ff.

in the domain of a remedy called the Leistungskondiktion All othercases

are classified under three other categories, of which the most notable is

the Eingriffskondiktion orenrichment claim based on encroachment.52

Forpresent purposes, the Leistungskondiktion is of particular importance.

According to the prevailing view, this remedy is available when

some-one has made a Leistung – that is, intentionally enlarged another’s estate

with a specific purpose in mind53– and this purpose for some reason oranotherthen failed As DieterReuterand Michael Martinek put it, ‘in

the case of the Leistungskondiktion the lack of the legal ground for the Leistung lies in the failure of its purpose’.54 Since the Leistungskondiktion

unites various cases of retention without legal ground brought about by

a transfer, it is not surprising that German authors have found it nient to categorise the fields of application of this remedy along the lines

conve-of the condictiones conve-of Roman law.55 According to the prevailing view, the

Leistungskondiktion in its condictio indebiti guise would then be applied when

the purpose of a transfer failed because the intended discharge of a debtdid not materialise.56Afterall, the condictio indebiti is the civil-law remedy par excellence forthe recovery of an indebitum, ortransferwhich is not due

or owed However, there is also a minority opinion, which regards the phasis on the purpose of the transfer as unnecessary To the supporters

em-of this view, the legal ground for a Leistung is simply the relationship em-of

indebtedness to which it pertains.57 A contract of sale, for example,

pro-vides such a relationship when the Leistung takes the form of a payment

of the purchase price Whether it was intended to achieve the purpose

of payment is not relevant if the relationship of indebtedness existed inany event Thus, if the purchaser in this example accidentally pays thepurchase price twice, but also happens to owe the seller a separate debt

52 On these various remedies, see Markesinis, Lorenz and Dannemann, 717 ff., 740 ff.,

752 ff.

53 See BGHZ 40, 272 (277); BGHZ 58, 184 (188).

54DieterReuterand Michael Martinek, Ungerechtfertigte Bereicherung (1983), 125.

55See DieterMedicus, B¨urgerliches Recht (18th edn, 1999), n 689; Reuterand Martinek, Ungerechtfertige Bereicherung, 125 ff.; Hans-Georg Koppensteiner and Ernst Kramer, Ungerechtfertigte Bereicherung (2nd edn, 1988), 49 ff.; for criticism see Markesinis,

Lorenz and Dannemann, 719.

56See Medicus, B¨urgerliches Recht, n 689; Koppensteinerand Kramer, Ungerechtfertige Bereicherung, 50; Reuterand Martinek, Ungerechtfertige Bereicherung, 126 ff.

57See Karl Larenz and Claus-Wilhelm Canaris, Lehrbuch des Schuldrechts (13th edn, 1994),

vol II/2§ 67 III 1 It might seem as if this approach is unable to deal with the datio ob rem, which is not supported by an obligation at the time of transfer However, such a datio is still supported by an underlying understanding as to the basis on which it is

to be held by the recipient This is regarded as sufficient to justify the retention of what has been given by the recipient.

(say, under an agreement of loan), the seller can set off the purchaser’senrichment claim based on the second payment against his own claim forpayment of the loan.58 The second payment would then no longerbe re-tained without legal ground This consequence cannot be explained with

an analysis based on the failure of the purpose of the second payment Infact, it indicates that such an analysis is unnecessary

(ii) Leistungskondiktion in cases of fraud and duress

To return to the original inquiry, namely to establish what enrichmentremedy should be used to recover transfers made where a contract wasrescinded due to fraud or duress As far as fraud is concerned, the appro-

priate remedy would be the Leistungskondiktion Underthe prevailing view,

it should be available because a transfer has been made with the purpose

of fulfilling an obligation undera contract, and the purpose failed cause the contract was rescinded And under the minority view, it should

be-be available simply be-because rescission removed the relationship of

indebt-edness or causa for the transfer Hence, in terms of both theories, the

transfer is not due, and is retained without legal ground, rendering it

recoverable with the Leistungskondiktion.59

In the case of duress, the situation is somewhat more complicated First,

it has to be determined whether a Leistung is being dealt with at all In

this regard, it is necessary to return briefly to a question raised in treatingthe content of duress, namely whether it ‘overbears’ the will or merely

‘deflects’ it If the ‘overborne will’ theory is followed, it seems as if the

relevant concept is an Eingriff or ‘encroachment’ by the wrongdoer, rather than with a Leistung as defined above Put in very simple terms, there

is no longera ‘giving’ but rathera ‘taking’ of the benefit If so, then

the Eingriffskondiktion would have to be awarded.60 However, as indicatedabove, the preferable approach is that duress does not destroy consent, butinstead improperly influences or ‘deflects’ it To what extent, then, would

the Leistungskondiktion be an appropriate remedy? Here the prevailing and

minority views have to be contrasted

According to the prevailing approach, the benefit had to be conferred

with a specific purpose in mind, and this purpose then had to fail The

problem is to determine what the purpose of the transfer was One sibility is that the purpose was to fulfil the obligations created by theextorted contract, and that this purpose then failed due to the rescission

pos-58 See§ 387 BGB.

59 See Lieb, in: M¨unchener Kommentar, § 812, nn 137 ff.

60 Du Plessis, Compulsion and Restitution, 72 ff., 159 ff.

The Leistungskondiktion in its condictio indebiti guise should then be available.

However, it seems rather artificial to say that a person who is so averse

to being bound that he has to be illegally forced to enter into a contractwould really have the purpose to fulfil obligations under such a contract

A more convincing view seems to be that the purpose of the transfer istemporarily to avert the harm which would flow from non-fulfilment Theproblem is that if the victim then succeeds in averting harm through mak-ing the transfer, its purpose has in fact been achieved, which means that it

is precisely the type of situation where the Leistungskondiktion, as ally understood, should not be awarded However, the minority approach

tradition-to the Leistungskondiktion completely avoids these problems: the mere fact

that the declarations of intent required for the conclusion of a contract

have been rescinded means that no relationship of indebtedness or causa

can exist Hence, as in the case of fraud, the retention is without legalground, and the question whether the ‘purpose’ of the transfer has beenfulfilled is irrelevant.61 Underthe minority view, the Leistungskondiktion

should therefore be available to reclaim the transfer made in fulfilment

of a contract concluded under duress

(iii) The approach of Dutch law

In modern Dutch law a different picture emerges Unlike a number ofothercivilian systems,62 it has consciously grouped together things that

are given without legal ground in art 6:203 BW This article, which is

dis-tinct from the general provision dealing with unjustified enrichment,63provides that all undue payments are to be restored, even though no shift

of wealth (in other words, no enrichment) took place Thus, even thoughthe ‘payment’ remained the property of the payer, it is still recoverable un-der this provision The reason is that Dutch law, contrary to German law,subscribes to a causal system of the transfer of property The purpose is tosimplify life forthe claimant All he has to do is prove that something notdue was given and received Proof of ownership is not required,64and loss

of enrichment or change of position is no defence either.65One instancewhere a payment is recoverable under this section is where an initially

61See Lieb, in: M¨unchener Kommentar, § 812, n 138.

62 See, e.g., on the German, Swiss and Greek codes, E J H Schrage, ‘Restitution in the

New Dutch Civil Code’, [1994] Restitution LR 208, 209 On the background to the code see C J van Zeben and J.W du Pon (eds.), Parlementaire Geschiedenis van het Nieuwe Burgerlijk Wetboek, vol VI, Algemeen Gedeelte van het Verbintenissenrecht (1981), 802 ff.

63 Art 6:212 NBW.

64Van Zeben and Du Pon, Parlementaire Geschiedenis, 803.

65 Schrage, ‘Restitution’, 209, n 8.

valid obligation is rescinded on grounds of fraud or duress.66 BecauseDutch law follows a causal system of transfer, there is also a concurrencewith vindicatory remedies, based on ownership.67The practical difference

is that the latter remedy can only be used to obtain restitution in natura

of a thing If it is lost ordamaged, the remedy is useless On the otherhand, in the case of insolvency, the owner can use his vindicatory remedy

to reclaim his property, and will not be left with a mere personal claim

(iv) The common-law approach

The crucial point in these civilian jurisdictions is that fraud and duressare merely grounds which indicate that a contract is invalid This rendersperformances under the contract undue, and (in German law at least)

results in the performances being retained without legal ground or sine causa Ultimately, relief could then be provided in unjustified enrichment.

But this is, of course, not the only way of viewing the function of fraudand duress It can be argued that it is not necessary to work with modelsthat employ concepts such as ‘undue’ or‘retention without legal ground’.Why not simply say that where a contract is concluded under fraud andduress, there should be restitution of any performance made in fulfil-ment of such a contract? Here the views of common lawyers who followthe ‘unjust factors’ approach to the law of restitution are of interest.68Since othercontributions deal with this approach at length, this is only

a brief overview of what unjust factors have to be relied on in claimingrestitution where a contract has been concluded under fraud or duress.This is not the easiest of tasks.69Forexample, fraud as such does not seem

to be an unjust factor (in fact, the word ‘fraud’ does not even feature in

66 Art 3:44 BW; Hartkamp and Tillema, Contract Law, §§ 102 ff., 328 Additional

provisions apply if the recipient is in bad faith: he then is, among other things, liable for all damage and payment of interest from the time the obligation arose.

67 See section III, 3, below On the relationship between these remedies see Scheltema,

Onverschuldigde betaling, 80 ff., 130 ff.

68 It should be noted that these factors are not determined by the law of contract, but

by the law of restitution itself It is obvious that the law of restitution, thus

conceived, has a much greater role to play in common-law systems than the law of unjustified enrichment in civilian systems As Brice Dickson puts it, ‘[r]ather than being a particular head of recovery with a small part to play in the law of obligations

it is supposedly a great unifying principle underlying the whole range of

restitutionary remedies’ (‘Unjust Enrichment Claims: A Comparative Overview’, [1995]

54 CLJ 100, 101; also see Steve Hedley, ‘Unjust Enrichment’, [1995] 54 CLJ 578).

69 See Andrew Burrows, The Law of Resitution (1993), 106 (he refers to the ‘quicksilver

requirement’) For criticism of the ‘unjust factors’ approach from a comparative perspective see Zimmermann, ‘Unjustified Enrichment’, 414.

the index of one of the basic textbooks).70Some argue that the appropriateunjust factoris in fact errorormistake, since, as seen above, all misrep-resentations induce error.71 However, the reprehensible manner in whichthe error has been induced in the case of fraud requires exceptional pro-tection for the victim Fraud therefore needs to be dealt with separatelyfrom other cases of error Others argue that in cases of mistake of fact (andpresumably also those cases where the mistake was induced by fraud), therestitutionary ground of ‘failure of consideration’ is more appropriate Theproblem here, though, is that there are different views on what this expres-sion is supposed to mean.72Anotherpossible unjust factor(more relevant

to situations not involving a contract) is ignorance Ignorance apparentlywould be the appropriate unjust factor where a third party obtained some-thing without knowledge of the claimant as a consequence of another’sfraud.73As forduress, it can be brought home underthe unjust factorof

‘compulsion’,74but fora civil lawyer(at least), its relationship with some

of the othercases of compulsion seems ratherremote.75 Finally, there isthe idea that the use of an improper method of obtaining consent couldamount to a ‘wrong’, although the exact relationship with compulsion as

an unjust factoris not clear: one will have to look at the way the judgedeals with the claim.76

70Burrows, Law of Restitution,

71See PeterBirks, An Introduction to the Law of Restitution (1985, revised edn, 1989), 167 ff.

72See Burrows, Law of Restitution, 108 ff On the idea that in the case where a contract

has been avoided on grounds of mistake (and presumably fraud), the unjust factor is

‘failure of assumptions’, see Andrew Tettenborn, The Law of Restitution in England and Ireland (2nd edn, 1996), 16, 135 ff., 139 On the dangers involved in introducing

‘absence of consideration’ as an unjust factor see Peter Birks, ‘No Consideration:

Restitution afterVoid Contracts’, (1993) 23 University of Western Australia LR 195, 231 ff.

(as quoted in Andrew Burrows and Ewan McKendrick, Cases and Materials on the Law

of Restitution (1997), 294 ff.; cf also the speech of Lord Goff of Chieveley in

Westdeutsche Landesbank Girozentrale v Islington LBC [1996] 2 All ER 961 (HL) at 967 ff.).

73Cf Transvaal and Delagoa Bay Investment Co v Atkinson [1944] 1 All ER 579, which deals

with the situation where the person committing the fraud knew that cheques were signed routinely.

74See Birks, Introduction 174–84 Burrows, Law of Restitution, 161, seems to regard the

relationship between duress and compulsion more loosely; he merely states that duress can be distinguished from ‘other autonomous unjust factors that, to a greater orlesserextent, also deal with compulsion, such as legal compulsion, necessity and

exploitation’ Tettenborn, Law of Restitution, 14–15, 77 ff avoids the generic concept

‘compulsion’ and simply regards duress as a ‘factor qualifying voluntariness’.

75See Du Plessis, Compulsion and Restitution, 210 ff The most notable example is the

treatment of ‘moral compulsion’ From a civilian perspective the idea that the good

Samaritan or negotiorum gestor acts undercompulsion seems ratherunusual.

76Birks, Introduction, 318 ff.

Altogetherthis is not a model that is manifestly clearoruser-friendly.However, if these difficulties of approach are disregarded for the mo-ment, and it is simply assumed that fraud and duress can generally beregarded as (at least) belonging under the unjust factors of ‘mistake’ and

‘compulsion’, one important concluding observation can be made It hasbeen said that these unjust factors can be brought home under the genericterm of ‘non-voluntariness’, which covers the situation where one persondid not intend anotherto have something It has also been said that this is

the field of the condictio marked out by Lord Mansfield in the century decision of Moses v Macferlan.77 As seen above, the Leistungskondik- tion in modern German law also covers the field of the condictio, through

eighteenth-unifying all cases where a person failed to achieve a certain purpose whenmaking a transfer to another Thus, it seems that on a very broad levelthere are certain similarities between organising concepts of the civil law

of unjustified enrichment and the common law of restitution

But there is one crucial difference The moment we ask ‘why was it not

intended that anothershould have it ?’ – in otherwords, the moment westart subdividing these categories – the responses vary dramatically Thecivil lawyer immediately responds that one important reason why it wasnot intended that anothershould have something is because it was notdue The case where a transfer was not due because a contract was re-scinded on grounds of fraud or duress just happens to be one of many,and the civilian enrichment lawyer is not interested in devising long lists

of all these cases This is not to say that fraud or duress are irrelevantwhen determining whether certain rules modifying the measure of recov-ery should apply, but this does not detract from their primary function

On the other hand, the response of some common lawyers to the samequestion would be that the reason why restitution should be provided

is that the victim was mistaken orcompelled To them the ‘due/undue’distinction simply masks underlying reasons for providing restitution –reasons which should be the legitimate concern of enrichment lawyers.However, for present purposes it is not necessary to deal at length withthese differences They are not only dealt with in greater theoretical de-tail by other contributions, but from a practical perspective they do notseem to matter very much: both ultimately result in enrichment-based

or restitutionary relief being provided where a contract was rescinded ongrounds of fraud or duress

77 PeterBirks, ‘Restitution: A View of Scots Law’, [1985] Current Legal Problems 57, 67 The reference is to Moses v Macferlan (1760) 2 Burr 1005 at 1009.

2 The effect of fraud and duress on the validity of a performance 78

Where a transfer has been made in fulfilment or performance of a tract concluded under fraud or duress, its recoverability is dealt with interms of the principles set out above However, as stated at the outset,not all transfers are aimed at the performance of contractual obligations.What impact then do fraud and duress have on the recoverability of thesetransfers? In the civil-law systems under review, the answer depends ontwo issues, namely whether the transfers were in fact due, and whetherachieving fulfilment or performance requires the type of act or declarationwhich can be rescinded

con-(a) Undue transfers: the condictio indebiti

It has been indicated above that one of the most important features of thecivil law of unjustified enrichment is that it awards an enrichment rem-edy if something was given that turned out not to be due This notion has

a certain rational appeal, for why, if something was not owed, should therecipient be entitled to retain it? However, a number of systems recognisethat the mere fact that something not due was given does not automati-cally entitle the transferor to an enrichment remedy It is not necessaryhere to recount the centuries-old debate on whether the claimant who

seeks relief with the condictio indebiti had to prove a factor (such as error) that justifies the departure of the indebitum from his hands, orwhetherit

is up to the recipient to argue that recovery should be barred because ofthe transferor’s knowledge that it was due.79

To focus solely on the modern systems under review, it is clear that theapproaches diverge considerably South African law, which has retained

the condictio indebiti in uncodified form, does not only require proof of an

undue transfer, but also further factors such as excusable error or certain

78On the nature of performance in German law in general see Joachim Gernhuber, Die Erf¨ullung und ihre Surrogate sowie das Erl¨oschen der Schuldverh¨altnisse aus anderen Gr¨unden

(2nd edn, 1994),§§ 5 ff.; Helmut Heinrichs, in: M¨unchener Kommentar zum B¨urgerlichen Gesetzbuch (3rd edn, 1994) vol II, § 362, n 2; Josef Esserand Hans-Leo Weyers, Schuldrecht (7th edn, 1991), vol II, 429 ff.; Karl Larenz, Lehrbuch des Schuldrechts (14th

edn, 1987), vol I,§ 18; for Dutch law, see arts 6:27–51; 6:111–26 BW; Hartkamp, Mr C Asser’s Handleiding, n 184 The effect of fraud on various methods of performance,

such as the Continental counterparts of negotiable instruments, documentary credits, or credit or charge cards, which could involve a host of a additional

(collateral) agreements, regulating the rights of third parties or stopping the process

of payment, cannot be considered here (see in the context of German law,

Gernhuber, Die Erf¨ullung, § 9 II; Larenz, Lehrbuch, § 18 IV).

79See Zimmermann, Law of Obligations, 834 ff., 848 ff., 866 ff.

forms of compulsion (under the influence of the English law of ‘paymentsmade underduress of goods’).80 In German law, the approach is more in-direct: proof of error or compulsion is not required, but (as seen above inthe case of a performance under a contract rescinded on grounds of fraud)knowledge on the part of the transferor that no debt is due could be raised

as a defence barring recovery The relevance of fraud is then to indicatethat the victim acted in error, and consequently could not have knownthat no debt was due It will further be recalled that the bar to recoverybased on knowledge could not apply in cases of duress: it cannot be saidthat the victim acted contrary to his previous conduct if such conductwas compelled.81In Dutch law, the protection of the transferor is greater:even knowledge that no debt is due does not bar recovery And yet, evenhere, the fact that the performance has been obtained through fraud andduress is not irrelevant Since it influences the question whether the recip-ient would be considered to be in bad faith, and hence in default, it maygive rise to increased liability.82All in all, it is clearthat outside the con-tractual context too, fraud and duress influence the relief which the law

of unjustified enrichment can give a person who made an undue transfer

(b) Due transfers

If a due transfer (irrespective of the source of the obligation) is obtainedthrough fraud or duress, the instinctive reaction is that there can be noliability based on unjustified enrichment, since there is no enrichment.Afterall, if the transfermerely takes the place of the obligation, the re-cipient’s net worth remains the same However, this would only be thecase if the transfer in fact succeeded in validly discharging the obligation.The question that therefore now arises is whether fraud or duress might

in fact influence the validity of performance To answer this question itneeds to be established whether performance involves some declaration

of intent oract that could be rescinded Unfortunately, in the civilian

80 See Commissioner for Inland Revenue v First National Bank Ltd 1990 (3) SA 641 (A) 646 ff.;

J G Lotz, ‘Enrichment’, in: W A Joubert (ed.), The Law of South Africa (1st reissue,

1996), vol IX,§ 79 On the problems surrounding the determination of the key requirements of the condictio indebiti in anothermixed jurisdiction, namely Scots law,

see Robin Evans-Jones, ‘Unjust Enrichment, Contract and the Third Reception of

Roman law in Scotland’, (1993) 109 LQR 643; also his ‘Some Reflections on the Condictio Indebiti in a Mixed Legal System’, (1994) 111 SALJ 759; Du Plessis, Compulsion and Restitution, 202 ff., 221 ff.

81 See section III, 1, (a) above.

82 See Schrage, [1994] Restitution LR 208, 215 and the treatment of arts 6:205, 6:84 ff BW.

systems underconsideration, no unanimity exists on this point Classical

Roman law knew the doctrine of solutio, but this simply entailed that you

did what you had obliged yourself to do.83There was no refined analysis

as to whether solutio is a unilateral or bilateral act, or with what intention

it has to be made In the ius commune, no particular doctrinal approach

prevailed either.84 It is not surprising then that the position in moderncivilian systems is anything but settled.85

Here only a brief review of the main approaches to the nature of formance will be attempted Particular emphasis will be placed on theposition in German law, where the debate has reached a considerable de-gree of sophistication First, there is the traditional approach that perfor-

per-mance requires agreement between the parties that a transfer is intended

to fulfil a specific obligation.86 Since such an agreement would requirevalid declarations of intent, fraud or duress could conceivably affect thevalidity of performance However, this approach has difficulties in accom-modating performance through rendering a service or omission, where

83See Ulp D 50, 16, 176; 12, 6, 63; 46, 3, 54; 46, 3, 80 pr.; J A C Thomas, Textbook of Roman Law (1976), 343; Zimmermann, Law of Obligations, 748 ff.

84 Most Roman-Dutch authors did not expressly require an agreement whereby

performance extinguishes the obligations; see Hugo de Groot, Inleidinge tot de Hollandsche Rechts-geleertheyd (ed Robert Warden Lee, 2nd edn, 1953), III, XXXIX, 8; Johannes Voet, Commentarius ad Pandectas (1827, 1829), XLVI, III, 1 (translated by Percival Gane, The Selective Voet, being the Commentary on the Pandects (1957), vol VII, 93).

In fact, it even was said that a third party could pay on behalf of an unaware or unwilling debtor (see ibid., XLVI, III, 1) However, the seventeenth-century German authorWolfgang Adam Lauterbach and, following in his footsteps, the

eighteenth-century Roman-Dutch author Willem Schorer seem to have believed that proper performance does require the consent of both parties (see Lauterbach,

Collegium theoretico-practicum (1723 ff.), Lib XLVI, Tit III, nn 3, 4; Willem Schorer, Aantekeningen over de inleydinge tot de Hollandsche rechtsgeleerdheid (translated into Dutch

by J E Austen, 1784–6), III, XXXIX, 7.

85 In fact, Wolfgang Fikentscher regards the legal nature of fulfilment as belonging to

the most contentious areas in the whole of the law of obligations: Schuldrecht (9th

edn, 1997),§ 268 The compilers of the German Civil Code were careful to avoid

making an explicit choice, and preferred to leave it in the hands of legal scholarship

to come up with an answer (see Mugdan, ‘Motive’, 81) On German law in general,

see Helmut Heinrichs, in: Palandt (ed.), B¨urgerliches Gesetzbuch (59th edn, 2000), § 362,

nn 5–7; Heinrichs, in: M¨unchener Kommentar, § 362, nn 5–14; Larenz, Lehrbuch,

§ 18 I In the law of the Netherlands, uncertainty also prevails (see Hartkamp, Mr C Asser’s Handleiding vol I, n 186).

86On this theory, which is known as the Vertragstheorie in German law, and which really

represents a continuation of the views which were popular in the last stages of the

ius commune, see Larenz, Lehrbuch, § 18 I sub 2; Heinrichs, in: M¨unchener Kommentar,

§ 362, n 6 with footnote 12; Gernhuber, Die Erf¨ullung, § 5 II 3; cf also RGZ 60, 24 (28).

In the Netherlands, such an approach was advanced by M H Bregstein, Ongegronde Vermogensvermeerdering (Ph.D thesis, 1927), 278.

it is rather artificial to work with the idea of ‘acceptance’.87 One mightargue (as modern South African law does88) that these cases are merelyexceptions to the rule, but this still leaves the problem that the creditormight refuse to accept performance An approach currently more popularmaintains that performance requires only that the debtor indicates thepurpose for which the benefit is conferred.89It is not quite clearwhetherthis determination of the purpose of the transfer is itself a juristic act.Apparently it is only aimed at ‘assigning’ the transfer to a debt, and not

at achieving the legal consequence of fulfilment.90However, it does seem

as if some declaration of will is made, which at least would be voidable

in cases of error.91 Although this approach also suffers from defects,92 itdoes have one notable advantage It will be recalled that, for purposes

of the German law of unjustified enrichment, a Leistung is defined as the

conscious conferment of a benefit with a certain purpose in mind Thisapproach adopts essentially the same definition for purposes of defining

Leistung in the context of performance, and therefore streamlines the

ap-plication of this concept in different branches of law However, this viewdoes not enjoy general acceptance

The approach which currently prevails in German law rejects any notionthat performance requires an ‘intention to fulfil’ or an indication as to the

‘purpose’ of a transfer All that is required is for a benefit to be conferred.93

In other words, the debtor simply has to bring about performance in a

87 Forotherdifficulties see Gernhuber, Die Erf¨ullung, § 5 II 3 On the rejection of this approach in Dutch law, see Hartkamp, Mr C Asser’s Handleiding, n 187.

88 See J C de Wet and A H van Wyk, Kontraktereg en Handelsreg (5th edn, 1992), vol I,

183, 263, n 73; Gerhard Lubbe and Christina Murray, Farlam & Hathaway, Contract: Cases, Materials and Commentary (1988), 716; Schalk van derMerwe et al., Contract: General Principles (1993), 359, referring to Thienhaus v Metje & Ziegler Ltd 1965 (3) SA 25 (A) 45 On related attempts at modification through the beschr¨ankte Vertragstheorie in German law see Larenz, Lehrbuch, § 18 I sub 3; Heinrichs, in: M¨unchener Kommentar,

§ 362, n 7 with footnote 13; Gernhuber, Die Erf¨ullung § 5 II 4.

89 On the Theorie der finalen Leistungsbewirkung see Larenz, Lehrbuch, § 18 I; Heinrichs, in: M¨unchener Kommentar, § 362, n 10 with footnote 16; Gernhuber, Die Erf¨ullung § 5 II 8.

90 Larenz, Lehrbuch, § 18 I; but see Hans Wieling, ‘Empf¨angerhorizont: Auslegung der

Zweckbestimmung und Eigentumserwerb’, 1977 JZ 291.

91 Cf DieterMedicus, Schuldrecht I: Allgemeiner Teil (11th edn, 1999), §23 IV 3 (he refers to

BGHZ 106, 163, 166); Heinrichs, in: Burgerliches Gesetzbuch, § 362, n 5.

92 The mere fact that an indication of the purpose of the transfer occasionally is important (e.g where there are several debts and it has to be determined which of them the debtorwants to fulfil) does not imply that such an indication always

should be required; see Heinrichs, in: M¨unchener Kommentar, § 362, n 13; Larenz, Lehrbuch, §18 I.

93 See Larenz, Lehrbuch, § 18 I; Medicus, Schuldrecht I, § 23 IV 3; Gernhuber, Die Erf¨ullung,

§ 5 II 6.

‘real’ or factual sense Thus, in the case of sale, performance would takeplace if ownership was transferred to the purchaser.94 The ‘allocation’or‘assignment’ of the transferto the fulfilment of a specific debt isautomatic.95 Unsurprisingly, this approach, too, has been subject to crit-icism It fails to show which obligation is to be fulfilled if there is aplurality of debts; it has also been doubted whether it is acceptable to de-

sign a theory around the exceptional case (such as omissio) where a purely

factual performance would indeed suffice, instead of the commoner case(such as delivery) where transfer does require some juristic act.96

In the light of the above, it should therefore be clear that the tentious nature of performance leaves the question unanswered whether

con-a due trcon-ansfer obtcon-ained through frcon-aud or duress is recovercon-able withenrichment-based remedies However, the question can be asked from apolicy perspective Should the victim merely be satisfied with a delictualclaim, based on the violation of his dignity and physical integrity, or withthe cold comfort of the criminal sanction? On the one hand, there is thepolicy objective that one should not waste the time of the courts by re-claiming what has to be transferred in any event This is encapsulated in

the ancient civilian maxim of dolo facit qui petit quod redditurus est, which essentially states that in such a case a person acts with dolus.97 But thismaxim makes no distinction as to the way in which the defendant ob-tained what now has to be returned Presumably it is aimed against thetype of ‘technical’ reliance on a right in a way that violates bona fides orgood faith.98 On the other hand, there is ample civilian authority whichsupports the policy objective that where a person takes the law into hisown hands, and disturbs the peace, he should be made to return what

he has obtained, irrespective of whether he is entitled to it or not Herethe relief which for centuries has been provided to possessors of prop-erty, regardless of the validity of their title, is particularly relevant.99 All

94§ 929 BGB The same holds true for cession: § 398 BGB (Abtretung).

95 Obviously, specific rules apply in the case of appropriation of performance; see§ 366

(1) BGB.

96Fikentscher, Schuldrecht, § 270. 97Paul D 44, 4, 8 pr.; D 50, 17, 173, 3.

98See Zimmermann, Law of Obligations, 668 The accompanying example in Paul D 44,

4, 8, 1 indicates a more restricted scope of application for the maxim It reads as follows: ‘[t]hus, if an heirhas been condemned not to claim from the debtor, the debtor can employ the defence of fraud, as well as bring an action based on the will’.

99See, from a South African perspective, the treatment of the mandament van spolie in

C G van derMerwe and M J de Waal, The Law of Things and Servitudes (1993), §§ 73 ff Although the roots of the mandament lie in Canon law, Roman law knew possessory

interdicts, which fulfilled related functions.

in all, the notion that enrichment-based remedies can be provided where

a due transfer is obtained through improper means therefore cannot bedismissed out of hand.100

On the assumption that such a claim should be allowed, it may be asked

in conclusion what civil-law remedy could most conveniently be used ward this end In the light of the fact that we are dealing with a due

to-transfer, any modern version of the condictio indebiti seems inappropriate.

Anotherpossibility is to have recourse to a long but almost forgottenline of civilian authority which regards the application of extortion as

so improper that it gives rise to illegality, and hence to another condictio, the condictio ob turpem vel iniustam causam However, the references in the ius commune to this remedy ultimately derive from a text dealing with

a transfer that is undue, because it was made pursuant to an extortedstipulation.101 In modern German law, one of the (last remaining) fields

of application of the Leistungskondiktion in its condictio ob turpem vel tam causam guise are in fact ‘extortion-related’ situations, but this again

inius-applies to payments which are not due.102 It may further be argued that

duress amounts to a type of Eingriff orencroachment, but (as pointed

out before) in cases of extortion we are still dealing with a Leistung or

‘giving’, rather than the type of ‘taking’ that an encroachment would quire Ultimately, the solution may well be to recognise a whole new field

re-of application re-of the Leistungskondiktion, based on retention re-of a transfer

without legal ground due to impropriety of the method used to obtain it.Such a development seems rather unlikely in the light of the approachcurrently favoured to the nature of performance in German law

As faras South African law is concerned, the condictiones indebiti and

ob turpem vel iniustam causam would be inappropriate for reasons similar

to those mentioned when dealing with the German law However, South

African law does recognise the condictio sine causa specialis In essence, it is

a residual category, covering cases of sine causa retention that do not fall

100 Cf the Sri Lankan case of Attorney-General v Saibo (1923) NLR 321, where a payment

made in consequence of an unlawful threat of detention of some boats by the customs authorities was held to be recoverable, even though it was due Incidentally,

in that case it was held that this position also conforms with the position in English

law; see Chief Justice Bertram’s discussion of Sowell v Champion (1838) 6 A & E 407;

112 ER 156; Clark v Woods (1848) 2 Ex 395; 154 ER 545 and Pitt v Coomes (1835) 2 A & E

459; 111 ER 178.

101 Pomp D 12, 5, 7.

102 See Larenz and Canaris, Lehrbuch, § 68 I 6; du Plessis, Compulsion and Restitution, 146 ff., but also see Attorney-General v Saibo (1923) NLR 321 at 324.

underthe established condictiones.103 In so faras the debt-extinguishingagreement required for performance can be rescinded, it can be argued

that we are dealing with a case of sine causa retention that warrants the

provision of relief by way of this enrichment remedy

3 The effect of fraud and duress on the validity of the transfer of ownership

It has been seen that fraud and duress may affect the ability of a transfer

to fulfil an obligation, depending on which theory of performance is lowed It is now necessary to go one step further and inquire whether fraud

fol-or duress might affect the validity of the transfer itself Here one ters the difficult problem of the relationship between the laws of prop-erty and unjustified enrichment In this regard the difference betweensystems that follow the abstract and causal approaches to the transfer

encoun-of ownership is fundamental This distinction cuts across the boundaries

of the civil-law/common-law division Forexample, German law and SouthAfrican law follow the abstract approach, while Dutch law and English lawregard the transfer of ownership as causal What are the implications forthe choice of remedy where a transfer has been obtained as a consequence

of fraud or duress?

According to the abstract system, the validity of a transfer is not affected

by the invalidity of any underlying obligation that the transfer is supposed

to fulfil In this regard German law draws a clear distinction between a

Verpflichtungsgesch¨aft – that is, a juristic act (such as a contract of sale) that gives rise to an obligation to transfer – and a Verf¨ugungsgesch¨aft – that is,

a juristic act by means of which a right (such as ownership) is directly fected (i.e transferred, encumbered, changed or nullified) The rescission

af-of a declaration af-of will under§ 123 BGB (the provision dealing with fraud

and duress) may influence the validity of both these juristic acts If

own-ership has passed (that is, the Verf¨ugungsgesch¨aft was valid), the claimant

would have to use remedies based on unjustified enrichment.104 tory or ownership-based remedies are not available if the transfer (andmore specifically the real agreement required for transfer) is valid How-

Vindica-ever, if the Verf¨ugungsgesch¨aft itself is invalid, ownership cannot pass, and

103 See B & H Engineering v First National Bank of South Africa Ltd 1995 (2) SA 279 (A); Du Plessis, Compulsion and Restitution, 223.

104 See section III, 1, (b) above As Dernburg so graphically pointed out, the function of the law of unjustified enrichment is to heal the wounds inflicted by the abstract

system (B¨urgerliches Recht (3rd edn, 1906), vol II/2, 677 ff as quoted by Zimmermann, Law of Obligations, 867).

the ownershould then be able to institute a vindicatory claim against therecipient in terms of§ 985 BGB This paragraph holds that an owner can

demand that a possessor return his property.105 The problem, of course,

is to determine whether someone who has transferred something underfraud or duress is still the owner The key to this problem lies in§ 935(1)

BGB.106In essence this provides that acquisition of ownership is not ble if something has been stolen from the owner, gone missing, or other-

possi-wise been lost (abhanden gekommen) Since one way in which something

can become ‘lost’ is through duress (but not fraud),107 the claimant canrely on this section of the code to prove that ownership has not passed,and that the possessor should return his property in terms of§ 985 BGB.

However, not all cases of duress are regarded as sufficiently serious to

affect the validity of a Verf¨ugungsgesch¨aft, and it is difficult to determine where the border lies According to the Bundesgerichtshof,108 one wouldhave to prove that there was duress through irresistible physical violenceorequivalent psychological compulsion.109

If one turns to the position in modern South African law, it ally seems very similar to that prevailing in German law: it also favoursthe abstract system,110and therefore excludes vindicatory remedies wherethere has been a transfer of ownership However, there has been some sup-port for a causal system, especially where something has been obtained

gener-as a consequence of fraud.111 This is no doubt due to the influence ofRoman-Dutch law, which maintained that valid title could not pass under

a contract induced by fraud, and that a rei vindicatio had to be used to

effect recovery.112 But these views have not received general recognition,

105 Cf DieterGiesen, BGB Allgemeiner Teil: Rechtsgesch¨aftslehre (1991), n 303; Esserand Weyers, Schuldrecht, § 49 III.

106 On this paragraph, see generally, Friedrich Quack, in: M¨unchener Kommentar zum B¨urgerlichen Gesetzbuch (3rd edn, 1997), vol III, §935 It does not apply to money, bearer instruments, or things sold by public auction (cf.§ 935(2) BGB).

107 See Quack, in: M¨unchener Kommentar, § 935, n 9.

108 BGHZ 4, 10 (34 ff.).

109 This view has been criticised by J ¨ urgen F Baur and Rolf St ¨ urner in their edition of

Fritz Baur, Lehrbuch des Sachenrechts (17th edn, 1999), 610–11 (they favoura broader approach to cover all situations of Drohung).

110 See the majority judgment in Commissioner of Customs & Excise v Randles, Brothers & Hudson Ltd 1941 AD 369; Trust Bank van Afrika Bpk v Western Bank Bpk en Andere NNO

1978 (4) SA 281 (A); Air-Kel (Edms) Bpk h/a Merkel Motors v Bodenstein en ’n ander 1980 (3)

SA 917 (A); Van derMerwe and De Waal, Law of Things, §§ 13, 166–8.

111 See Kopelowitz v West & Others 1954 (4) SA 296 (W) 300; Van derMerwe and De Waal, Law of Things, § 167.

112 Fora critical assessment see J Scholtens, ‘Justa Causa Traditionis’, (1957) 74 SALJ 280,

285 ff.

especially since they seem to reflect some confusion between the ent effects of voidness and voidability on the validity of a contract itself

differ-as opposed to the real agreement needed to effect delivery under thecontract.113 It is only if fraud (and presumably duress) affect the validity

of the real agreement that a rei vindicatio may be awar ded.114

Dutch law subscribes to the causal system, which means that there is

no transfer of ownership if the causa is invalid As seen above, a curiousfeature of this system is that the transfer can be recovered not only with

a modern version of that central remedy available to the owner, the rei vindicatio, but also with a specific remedy used for the recovery of undue

payments.115 The relationship between these remedies is somewhat plex: in contrast to German law, some overlapping seems inevitable,116butthere are important cases where only the one remedy would be available.Forexample, only the claim based on undue payment would be available if

com-a pcom-ayment of money ccom-an no longer be identified Furthermore, if there hcom-asbeen a disposition to a third party, the victim will have to use the vindica-tory action against the third party If the third party is protected because

he was bona fide, the victim will only be left with an action based onundue payment against the recipient.117It is also important to stress thatthe sections of the Burgerlijk Wetboek that deal with recovery of unduepayments do not form part of the provisions relating to liability based onunjustified enrichment One is therefore dealing with something more

complex than the terrain usually covered by the condictio indebiti.118

4 Fraud and duress as delicts

The laws of delict and unjustified enrichment are both aimed at ing imbalances In the case of delict, the correction takes the form of anaward of damages as compensation for a civil injury or harm, while in thecase of the law of unjustified enrichment, it takes the form of a restitu-tionary award, aimed at taking away or ‘skimming off’ the enrichment.119

correct-113 See ibid., 288 ff.

114 See the analysis of Dalrymple, Frank & Feinstein v Friedman and Another (2) 1954 (4) SA

649 (W) at 664 provided by C G van der Merwe, Sakereg (2nd edn, 1989), 312, and approved of in The Tao Men 1996 (1) SA 559 (CPD) Also see Du Plessis, Compulsion and Restitution, 121, n 65.

115 See Scheltema, Onverschuldigde betaling, 130 ff on the relationship between arts.

Defining how exactly transfers obtained through fraud or duress should bedealt with undersuch a division has neverbeen simple There is a strand

of civilian authority which, out of rather overenthusiastic loyalty to the

coactus volui principle, regards the relief in the case of duress as delictual.

In essence, the argument is that what is willed under duress is none theless willed; duress would therefore not affect the validity of a contract,

and would not give rise to the remedy of restitutio in integrum.120 Thismeans that it would be up to the law of delict to effect restitution of thetransferred object.121However, civilian systems nowadays avoid such an ex-clusionary approach, and prefer to award delictual relief in conjunctionwith enrichment-based or contractual remedies This gives rise to someuncertainty with regard to concurrence of actions In South African law,

for example, the measure of the duty to provide full restitution (restitutio

in integrum) is close122to the measure of damages in cases of delict, wherethe aggrieved party has to be placed in the position in which he wouldhave been if the delict had not been committed.123 The German law ofduress also recognises delictual relief A person who is entitled to rescind

a declaration of will through Anfechtung is not only entitled to cancel a contract by relying on culpa in contrahendo,124 but can also claim dam-ages because of blameworthy conduct during negotiations.125 The claimfordamages is subject to § 249 BGB, which aims at putting the threat-

ened person in the position he would have been in, had someone not

120 See James Gordley, ‘Natural Law Origins of the Common Law of Contract’, in: John

Barton (ed.), Towards a General Law of Contract (1990), 398 Some Roman law textbooks also classify metus and dolus as praetorian delicts (see Thomas, Textbook, 373 ff.).

121 See James Dalrymple, 1st Viscount Stair, Institutions of the Law of Scotland (tercentenary edn by D M Walker, 1981), Book I, Title 9, 4; Du Plessis, Compulsion and Restitution,

55 ff.

122 But obviously not identical; see Van derMerwe et al., Contract, 106 Cf also M A Lambiris, Orders of Specific Performance and Restitutio in Integrum in South African Law (1989), 333, n 60 Forexample, restitutio in integrum does not coverconsequential loss.

123 Cf Davidson v Bonafede 1981 (2) SA 501 (C), where it is acknowledged that delictual

relief might take the form of ‘restitution’ of the purchase price According to Marais

AJ, ‘in the end, it seems to me to be a question of degree as to whether any

particular financial adjustment which falls to be made is one which is an integral

element in the granting of restitutio in integrum, oris one which is collateral to it, and

so should form the subject of a distinct claim for damages’ (at 511G) On the

relationship between restitutio in integrum and enrichment remedies in South African

law, see Visser, ‘Rethinking Unjustified Enrichment’, 215 ff.

124 Jauernig, B¨urgerliches Gesetzbuch, § 123, n 5; Giesen, BGB, nn 303, 307 But cf Hans Brox, Allgemeiner Teil des BGB (20th edn, 1996), nn 425, 413.

125 On culpa in contrahendo in general, cf Larenz, Lehrbuch, § 9 I; Friedrich Kessler and Edith Fine, ‘Culpa in Contrahendo, Bargaining in Good Faith, and Freedom of Contract:

A Comparative Study’, (1964) 77 Harvard LR 401 ff.

influenced his ability to decide freely.126Furthermore, if the duress is ered by the provisions of the criminal code dealing with coercion (§ 240 StGB – N¨otigung) and extortion ( § 253 StGB – Erpressung), it would also be

cov-possible to claim delictual damages under § 823(2) BGB, which provides

that a person incurs delictual liability if he infringes a statute intended forthe protection of others Finally, a delictual claim may be brought under

§ 826 BGB, which provides that a person who intentionally causes damage

to another in a manner contrary to morality is bound to compensate thatperson for the damage.127This also seems to be the position in Dutch law,which imposes delictual liability if an act was contrary to proper socialconduct.128

IV Conclusions

Although civil-law systems have traditionally recognised that something

obtained through dolus or metus (loosely translated as fraud or duress) has

to be restored, they have not been particularly consistent or clear as tothe exact legal basis of this obligation None the less, it would seem thatthe modern tendency is to base the duty primarily on the need to preventunjustified enrichment (rather than on contract or sui generis grounds)and to provide complementary or supplementary relief through the laws

of property and delict

If the conclusions of this chapter are restricted to the law of fied enrichment, it is apparent that fraud or duress are relevant in thefollowing contexts First, through influencing the validity of certain acts,they can remove the legal ground or basis for the retention of a benefit ortransfer The classic case is where a contract is invalidated through rescis-sion In some systems, it is specifically recognised that in such a case thepurported performance is retained without legal ground because it is no

unjusti-longerdue orowing – in otherwords, because it is an indebitum However,

the mere fact that a transfer is not due does not automatically give rise

to relief based on unjustified enrichment There are situations where thecircumstances surrounding the making of the undue transfer are suchthat its recovery cannot be allowed This leads to the second function offraud and duress

126 On the concurrence between the various claims, cf Giesen, BGB, n 307.

127 Cf ibid., n 303 and Medicus, B¨urgerliches Recht, n 626; Brox, Allgemeiner Teil, n 425 Generally see Larenz and Canaris, Lehrbuch, § 68 I 1 c).

128 See Hartkamp and Tillema, Contract Law, § 105; especially on the obligation to pay

damages, arts 6:162, 6:95 ff BW.

There is a long-standing controversy as to whether the person who made

an undue transfer must prove a further ground (most notably error, butalso compulsion) in orderto succeed, orwhetherit is up to the defen-dant to raise a defence (based usually on the knowledge that no debtwas due), which indicates that it was intended that the recipient couldretain the undue transfer Whatever the preferable approach may be, thefact patterns of fraud and duress are also relevant in this context, in thatthey assist in determining whether enrichment-based liability should beimposed Fraud fulfils this function by indicating that someone acted inerror (that is, wrongly believed a debt was due), while duress shows thatthere was no indication that a transfer could be retained They thereforeshow that, in both instances, the victim’s frame of mind was of such anature that relief should not be barred

However, there is a further related but distinct function of these proper methods of obtaining consent Even when it has been determinedthat enrichment liability should be imposed in principle, it may still influ-ence the exact measure of such liability While some civilian systems gener-ally prefer to hold the recipient liable only to the extent that he is still en-riched, it stands to reason that such protection should not be accorded tothe recipient who used fraud or duress to obtain the transfer His unaccept-able conduct and lack of bona fide belief in being entitled to retain what

im-he received should render him liable for tim-he full value of what im-he received.Three final observations may be made The first is that it is simplistic tostate that civilian systems only regard the fact patterns covered by fraudand duress as relevant so far as they indicate that contracts are invalid,and that contractual performances are therefore retained without legalground While fraud or duress may not be ‘unjust factors’ directly regarded

as grounds for recovery in unjustified enrichment, they certainly matterfor the rules regulating the restriction and extent of enrichment-basedrelief

The second observation is that, although the civilian tradition ally places strong emphasis on the recoverability of undue transfers, thepossibility cannot be discounted that fraud or duress could also renderdue transfers recoverable Much depends on whether rescission of perfor-mance of the due transfer is possible from a doctrinal perspective, andwhether policy considerations in any event justify providing relief If this

gener-is so, then fraud and duress could indeed be regarded as ‘factors’ thatjustify enrichment-based relief

The last observation is that if there is one aspect of the treatment offraud and duress in the civil law which deserves criticism, then it is the

lack of clarity surrounding the essential features of some of the specificenrichment remedies The prevailing view in German law regarding the

nature of the Leistungskondiktion is particularly problematic Although it

is said that this remedy should be awarded where a transfer has beenobtained through duress, it is not clear what the purpose of such a transfer

is supposed to be, and how this purpose is supposed to fail These problems

are avoided by minority approaches to the Leistungskondiktion, as well as

by Dutch law, which do not focus on any purpose of a transfer and simplyprovide relief based on the absence of a legal ground

part v

Change of position

225

226

8 Restitution without enrichment? Change of

position and Wegfall der Bereicherung James Gordley

I Introduction

In German law,1 American law2 and, more recently, English law, it is adefence to an action for unjust enrichment that the defendant is no longerenriched.3Nevertheless, many German scholars want to limit this defence

My former teacher, John Dawson, thought it leads to senseless results inGermany,4 which American courts avoid only by refusing to apply it Inthe United States, he said, ‘we would not as in Germany, conceive ofenrichment as a variable that can be recovered only as long as it lasts’.5

I do not like to quarrel with Dawson I spent my early professional lifebelieving he was infallible This once, however, he may have been mis-taken At any rate, I do not think this defence leads to senseless results aslong as it is confined to its original scope, using it to resolve the problems

it was originally meant to resolve

II The original scope of the doctrine

These problems become clear if the origin of the doctrine is examined Thedrafters of the German Civil Code took the doctrine from the nineteenth-century pandectists, Windscheid and Savigny Savigny seems to have taken

it from members of the seventeenth- and eighteenth-century natural law

1 § 818(3) BGB.

2 Restatement of the Law of Restitution, Quasi Contracts and Constructive Trusts (1937),

§ 142.

3 Lipkin Gorman (a Firm) v Karpnale Ltd [1991] 2 AC 548.

4 John P Dawson, ‘Erasable Enrichment in German Law’, (1981) 61 Boston University LR

271–314.

5 John P Dawson, ‘Restitution Without Enrichment’, (1981) 61 Boston University LR 563,

564.

school such as Grotius and Pufendorf They took it from a group centred

in Spain in the sixteenth century and known to historians as the ‘latescholastics’ The late scholastics had been discussing the implications ofAristotle’s concept of commutative justice as it had been interpreted byThomas Aquinas As Robert Feenstra has pointed out,6 their efforts gaverise to the modern idea of unjust enrichment as a separate body of lawcoeval with contract and tort

According to Aristotle, while distributive justice gives each citizen afair share of whatever resources a community has to divide, commuta-tive justice preserves each person’s share In involuntary transactions, onewho took or destroyed another’s resources has to give back an equivalentamount In voluntary transactions, parties have to exchange resources

of equivalent value.7 This distinction between involuntary and voluntarytransactions not only resembles the one now drawn between tort and con-tract, but seems to have been its lineal ancestor Our distinction goes back

to Gaius.8 Modern scholars believe that he took it from Aristotle.9

In any event, Thomas Aquinas explained that when one person hadacquired or interfered with another’s property, he might be liable for twodifferent reasons First, he might be liable because of the way in which he

did so (acceptio rei): he might have acted wrongfully, against the owner’s

will, in which case he was liable whether or not he still had the property;

or he might have acted with the owner’s consent, in which case whether

he was liable depended on the kind of voluntary agreement they hadmade.10 Secondly, he might be liable merely because he had another’s

property, regardless of how he had come by it (ipsa res accepta) According

to Aquinas, commutative justice required that he give it back.11

In this last case, according to the late scholastics, and then Grotiusand Pufendorf, a person who no longer has another’s property should

6 Robert Feenstra, ‘Grotius’ Doctrine of Unjust Enrichment as a Source of Obligation: Its

Influence on Roman-Dutch Law’ in: E J H Schrage (ed.), Unjust Enrichment (1995), 197.

7 Nicomachean Ethics, V, 9, 1130b–1131a. 8 Gai 3, 88.

9 Reinhard Zimmermann, The Law of Obligations: Roman Foundations of the Civilian Tradition (1990, paperback edn, 1996), 10–11; Max Kaser, Das R¨omische Privatrecht (2nd edn, 1971), vol I, 522; Anthony Honor ´e, Gaius (1962), 100; Helmut Coing, ‘Zum Einfluß der

Philosophie des Aristoteles auf die Entwicklung des r ¨omischen Rechts’, (1952) 69

Zeitschrift der Savigny-Stiftung f¨ur Rechtsgeschichte (Romanistische Abteilung) 24, 37–8.

10 Summa theologiae II–II, Quaest LXII, art 6 According to Aquinas, if the transaction was

purely for the benefit of the person who received the property – for example, a gratuitous loan – then compensation is due even if the property has been lost; if it was purely for the benefit of the owner – for example, a deposit – then compensation

is not due except if the loss was caused by grave fault.

11 Summa theologiae II–II, Quaest LXII, art 6.

still be liable if he has become richer by having once had it.12 Such aperson is liable only to the extent that he is still enriched Thus he isnot liable if he consumed another’s property or gave it away13 except tothe extent that he saved money he would otherwise have spent He is notliable if he bought and then resold another’s property except if he made aprofit.14

They reached this conclusion by first setting aside every other reasonthat the plaintiff might recover until all that is left is the defendant’senrichment by means of the plaintiff’s resources It is a defence that thedefendant is no longer enriched, but only if he no longer had the plaintiff’sproperty and is not liable because of the way he had initially acquired it,whether wrongfully or with the plaintiff’s consent The thesis of this chap-ter is that the doctrine is correct provided that it is confined to its originalscope It should be applied when the only reason the plaintiff should re-cover is that he has been enriched out of the defendant’s resources

III Cases outside the proper scope of the doctrine

The trouble in Germany arose, I believe, by oversimplifying this approach:

by thinking that a plaintiff who can neither find his property in the fendant’s hands, nor recover in tort or contract, must recover in unjustenrichment, and if so, can recover only to the extent that the defendant

de-is enriched German courts did so because of the structure of their code.They supposed that the plaintiff must make out his claim under§§ 812–22

BGB, which govern unjust enrichment, unless he can reclaim his propertyunder § 985 BGB, the modern version of the rei vindicatio, or recover in

tort, or recover in contract At that point, they either had to live with

the consequences of the doctrine of Wegfall or invent some excuse for

not applying it But could it not be that the reasons for allowing the

12Luis de Molina, De iustitia et iure tractatus (Venice, 1614), disp 718, n 2; Leonard Lessius, De iustitia et iure ceterisque virtutibus cardinalis (Paris, 1628), lib II, cap 14, dubitatio IV, n 3; Hugo Grotius, De iure belli ac pacis libri tres (Amsterdam, 1646), II, X,

2, 1; Samuel Pufendorf, De iure naturae ac gentium libri octo (Amsterdam, 1688), IV, XIII,

6 For a defence of this principle, see James Gordley, ‘The Principle against Unjustified

Enrichment’ in: K Luig, H Schack and H Wiedemann (eds.), Ged¨achtnisschrift f¨ur Alexander L¨uderitz (2000), 13.

13Lessius, De iustitia, lib II, cap 14, dubitatio I, n 5 (‘For example, if he spent ten gold

pieces of another’s property and only saved five of his own because he would only have consumed five otherwise, he is only liable for five because he appears to have become

richer only to that extent’); Grotius, De iure, II, X, 5; Pufendorf, De iure, IV, XIII, 9.

14Molina, De iustitia, disp 718, n 2; disp 721, n 6; Lessius, De iustitia, lib II, cap 14, dubitatio I, n 4; Grotius, De iure, II, X, 8; Pufendorf, De iure, IV, XIII, 8.

plaintiff to reclaim his property from a bona fide purchaser extend yond the case governed by § 985 BGB in which the purchaser still has

be-the property? Might it not matter whebe-ther be-the one party wronged or jured the other even if the plaintiff’s action is not in tort? Might it notmatter that a party did consent even if he did not enter into an enforce-able contract? If so, then the oversimplification leads in the wrong direc-tion, to a consideration merely of whether the defendant was enriched.What should be considered instead is the responsibility of a bona fide pur-chaser, or of a person who commits a wrong or injury, or of a person whomakes a voluntary decision even if it does not result in an enforceablecontract

in-1 The bona fide purchaser

In Germany, as elsewhere, an owner can reclaim his stolen car from a bonafide purchaser The reason, we now recognise, is not simply that the thiefhad no title to give It concerns the buyer’s ability to protect himself bydealing with someone who is reputable or at least amenable to a lawsuit

It may be that, for the same reason, the owner should recover from a bonafide purchaser who has resold the car If so, all one needs to say to justifythis result is to say that the bona fide purchaser can still sue whoeveroriginally sold the car to him German courts, however, see the remedy

in unjust enrichment, and then explain away the requirement that thedefendant be enriched by claiming that the money he paid the thief wasnot causally related to his enrichment.15As Dawson has pointed out, thatclaim is inconsistent with the loose approach they take to causation inother cases.16 As Werner Lorenz has observed, it is not helpful because itobscures the reason the plaintiff is allowed to recover, which has nothing

to do with whether the defendant is enriched.17

2 Wrong or injury

Suppose, next, that one of the parties committed a wrong or injury Fourtypes of cases are examined below: (a) the defendant wrongfully inducedthe plaintiff to contract, (b) the defendant exploited the plaintiff’s need

or ignorance to contract on unfairly advantageous terms, (c) the dant wrongfully appropriated a benefit that he should have obtained by

defen-15 BGHZ 14, 7, 9–10; BGHZ 9, 333, 335–6; BGHZ 55, 176, 180 (two bulls sold to butcher who cut them up, thereby acquiring title).

16 As noted by Dawson, ‘Erasable Enrichment’, 294.

17 Werner Lorenz, in: J von Staudingers Kommentar zum B¨urgerlichen Gesetzbuch (13th edn,

1994),§ 818, n 38.

contract, and (d) one party injured the other by disavowing the contractafter he had changed his position in reliance upon it, and yet the partycommitting the injury is not at fault because he is a minor, an insaneperson, or otherwise incompetent.

(a) Wrongfully inducing the other party to contract

Suppose the defendant fraudulently induced the plaintiff to buy a car,and that the car has been destroyed Lord Denning once said in a fraudcase,18 and I have argued elsewhere,19 that a person who commits anintentional wrong should be liable for its unforeseen consequences If so,then the reason that the plaintiff should recover the purchase price isbecause the defendant committed a wrong, not because he was enriched.German courts have reached this result by saying that the remedy is inunjust enrichment, but then creating an exception to the requirementthat the defendant be enriched Rather than entering by one door only toexit by another, it would be better to say that the remedy is not one forunjust enrichment

(b) Unfair terms

In another type of case, the wrong was to obtain unfair terms by ing the plaintiff’s ignorance or necessity The plaintiff cannot then claimthat, but for the wrong, he would not have contracted at all He wouldhave done so but on better terms The appropriate remedy, it would seem,would be to enforce the contract but on fair terms, which, in fact, is done

exploit-in the United States.20 If this result were thought to be too hard on a fendant who might not have contracted at all on such terms, he could begiven the option to reject these terms and rescind the contract That, in

de-fact, was done under the ius commune when a remedy was given for laesio enormis or gross disparity in price.21German courts face the difficulty that

§ 138 BGB prescribes only one remedy when a contract is one-sided: the

18Doyle v Olby [1969] 2 All ER 199 at 122 (CA).

19 James Gordley, ‘Responsibility in Crime, Tort and Contract for the Unforeseeable Consequences of an Intentional Wrong: A Once and Future Rule?’, in: P Cane and J.

Stapleton (eds.), The Law of Obligations: Essays in Celebration of John Fleming (1998),

175–208.

20 Uniform Commercial Code§ 2-302 See Jones v Star Credit Corp 298 NYS 2d 264 (SupCt 1969); Frostifresh v Reynoso 274 NYS 2d 757 (SupCt 1966), reversed as to damages, 281 NYS 2d 964 (App 1967); American Home Improvement Co v MacIver, 201 A 2d 886 (NH,

1964).

21James Gordley, ‘Just Price’, in: P Newman (ed.), The New Palgrave Dictionary of Economics and the Law (1998), vol II, 410.

contract is void.22Consequently, they apply the law of unjust enrichment

to return the parties, so far as possible, to their original position In onecase, a seller had been forced by his necessity to sell at too low a price, andthe buyer had resold at a profit The seller later became insolvent Thecourt awarded the seller the difference between the resale price and theinitial purchase price.23 Supposedly, that was the amount by whichthe buyer had been enriched Dawson believes that the seller should re-cover the initial purchase price and that the buyer should have to line

up with the seller’s other creditors to recover anything at all.24If, as gested here, the seller should merely recover the difference between a fairprice and the unfair purchase price, then both approaches are wrong Onesource of the difficulty is the rigidity of§ 138 BGB But another is the as-

sug-sumption of both the German courts and Dawson that the problem is one

of unjust enrichment rather than of devising a remedy for a certain type ofwrong

(c) Wrongful appropriation of a benefit which one should have obtained by contract

In another type of case, the wrong was knowingly to appropriate thing instead of obtaining it by contract Here, German courts not only

some-do not apply the some-doctrine of Wegfall but they some-do not even require that

the defendant ever have been enriched In a famous case, the defendantmanaged to take a trip to New York (and back again, when he was re-fused entry) without paying the plaintiff airline for his ticket The courtrequired him to pay even though he was not enriched by that amount.25When a railroad that put more traffic across a plaintiff ’s land than itsright-of-way, the court required it to pay the amount it would have had

to pay had it acquired the right in the normal way The railroad was notallowed to say that it would not have done so, had it expected to pay thatamount.26A motorcycle manufacturer that used a picture of a well-knownactor in its advertising, taken without his consent, was made to pay thelicence fee that would usually be paid to a celebrity.27 American courtshave reached similar results by holding that a contract has been formed

22 See generally Reinhard Zimmermann, Richterliches Moderationsrecht oder Totalnichtigkeit ?

According to the Second Restatement of Contracts, the offeree’s ‘silenceand inaction operate as an acceptance’ where he ‘takes the benefit of of-fered services with reasonable opportunity to reject them and reason toknow that they were offered with the expectation of compensation’.28Forexample, a person who reads newspapers delivered to him without hisrequest is deemed to have bought them even if he clearly did not intend

to do so.29 In such cases, to say a contract was formed is a fiction sincethe defendant did not consent Nevertheless, they are not cases of unjustenrichment in the ordinary sense since it does not matter whether thedefendant was enriched He is liable because he wrongfully attempted toappropriate a benefit instead of contracting for it He is not allowed tobetter his position by refusing to contract.30

(d) Injury caused by an incompetent party

A related question is what to do when an incompetent party has injuredsomeone by entering into a contract with a person who changes his posi-tion in reliance on it The competent party cannot ask a court to enforcethe contract and thereby protect his right to the bargain that he has made

He has not made a valid bargain because the incompetent party cannotgive consent Nevertheless, should he be able to recover any loss he hassuffered from the incompetent party? The question is like the one in tortlaw about whether insane people or minors should have to pay for thephysical damage that they do The answer is not obvious On the one hand,they are not at fault On the other, because of their condition, they havecaused someone else a loss Not surprisingly, different legal systems havegiven different answers although the most frequent approach is that theyare held liable in tort In the United States, the insane are liable in mostjurisdictions31 but not all.32 Children are not liable if they used the care

to be expected of a child of similar age.33 In French law until recently

28 Restatement (Second) of Contracts,§ 69 (1981).

29Austin v Burge, 137 SW 618 (Mo App 1911).

30 On the damages awarded in such cases, see James Gordley, ‘The Purpose of Awarding

Restitutionary Damages: A Reply to Professor Weinrib’, (2000) 1 Theoretical Inquiries in Law 39, 55–7.

31 Restatement (Second) of Torts,§ 283B (1963); W Page Keeton, Dan B Dobbs, Robert E Keeton and David G Owen, Prosser and Keeton on the Law of Torts (5th edn, 1984), 176–8;

Stephanie I Splane, ‘Tort Liability of the Mentally Ill in Negligence Actions’, (1983) 93

Yale LJ 153, 155–6.

32E.g., Breunig v American Family Ins Co 173 NW 2d 619 (Wis 1970).

33 Restatement (Second) of Torts,§ 283A (1963); Keeton et al., Prosser and Keeton on Torts,

179–82.

neither children nor the insane were held liable They are today They aresaid to be at fault by an ‘objective’ standard even for actions they couldnot help.34 In Germany and Italy, they are not liable on account of faultbut special provisions allow the court to award damages that are ‘equi-table’ considering the financial resources of the parties.35Thus, where anincompetent has caused a loss by entering into a contract, one might ex-pect legal systems to differ but that, most often, they would have to bear

it themselves

In the United States and Germany, however, the problem is approached

as one of unjust enrichment Yet, this approach has not preventedAmerican and German courts from protecting, albeit only partially, a partywho has suffered a loss by contracting with an incompetent In the UnitedStates, sometimes the contract is enforced, and, when it is not, the right

of the incompetent to recover for unjust enrichment is conditioned onthe return of whatever benefits he received.36

In Germany, protection was traditionally given by applying the so-called

Saldotheorie The claims in unjust enrichment of each party are aggregated

so that the defendant may deduct the value of whatever he gave the tiff from the value of whatever the plaintiff had given him For example,

plain-in one case plain-in which the doctrplain-ine was first applied, the plain-incompetent party,who was supervising the construction of a building, purchased steel gird-ers in his own name They were incorporated into the building which laterwas seized and sold to satisfy the demands of creditors After he died, hisheirs sued to recover the purchase price in unjust enrichment They recov-ered an insignificant amount because the court allowed the defendant todeduct the value of the girders.37 Dawson pointed out that if one wants

an incompetent person to bear a loss that he occasioned, this is hardly alogical way to do so.38The competent party is protected only when he is

a defendant He is protected if the incompetent party seeks money ages but not if title to a moveable has not passed, and the incompetentparty recovers it by self-help or without suing in unjust enrichment to

dam-do so Dawson also claimed that to give even this limited protection was

inconsistent with the doctrine of Wegfall, since the disappearance of the

34 The insane became liable when a 1968 statute changed the law Law of 3 January 1968,

now art 489(2), code civil See Patrizia Petrelli, ‘La responsabilit`a civile dell’infermo di mente nell’ordinamento francese’, (1991) 37 Rivista di diritto civile 77–86 Children were

held to the standard of care of a child of their own age until the decision of 9 May

1984, by the Cour de cassation meeting in assembl ´ee pl ´eni`ere: DS 1984.525 See Henri Mazeaud, ‘La “faute objective” et la responsabilit ´e’, DS 1985 Chron.13 at 86–95.

35 §§ 827–9 BGB; arts 2046–7, codice civile. 36 Dawson,‘Erasable Enrichment’, 298, n 91.

37 RG GruchB 55, 963. 38 Dawson, ‘Erasable Enrichment’, 299.

plaintiff’s enrichment – for example, by the incorporation of the girders

in the building – is ignored.39 He is quite right that the real question iswhether the competent party should be protected against the loss thatthe incompetent one would otherwise cause him But that is why the realbasis for liability is not that the incompetent has been enriched but ratherthat the competent should be protected Since it does not matter whetherthe incompetent was enriched, there is no reason why the doctrine of

Wegfall should apply.

3 Consent

A contract may be void, and yet one of the parties may still have made

a decision voluntarily which should affect his rights Examined below are(a) the decision to enter into a contract that turns out to be void and(b) the decision to employ a third party whose actions have prejudicedthe other party to the void contract

(a) The decision to enter into a contract that proves to be voidSuppose that the defendant has received the plaintiff’s property voluntar-ily, even though the transaction is not an enforceable contract: for exam-ple, he buys the plaintiff’s goods but the contract is void for defect in form.Suppose that these goods are now valueless because they were destroyed

or because they were specially made for the defendant, who does not wantthem any more If the contract is void, it follows that neither party canclaim the benefit of his bargain It does not follow that the transactionmust be regarded as involuntary for all purposes It might be that the risksshould still fall where they do whenever resources are exchanged volun-tarily If the goods are destroyed by chance, the risk falls on the buyer; ifthey are destroyed because they were defective when delivered, the riskfalls on the seller; if they were specially made to the buyer’s order, therisk falls on him that neither he nor anyone else will want them If so,then these results are appropriate because the reason for declaring thecontract void does not extend to all the consequences a voluntary trans-action normally carries with it Once again, they have nothing to do withwhether the defendant is enriched

Again, German courts arrive at these results, but to do so they find away around the requirement that the defendant must be enriched If thegoods were destroyed by chance after delivery, German courts again apply

39Ibid., 298.

the Saldotheorie The claims of buyer and seller in unjust enrichment are

aggregated so that the seller returns the purchase price less the value

of the goods he had delivered and which the buyer is unable to return

to him Thus, while the risk of destruction falls on the buyer, the sellerdoes not receive the benefit of his bargain, as he would if the contractwere enforceable If the goods are worthless because they were defective,

the Saldotheorie has been construed to allow the buyer to recover the

pur-chase price,40 an explanation that German authors regard as artificialeven though they approve of the result.41 In one case, where goods werespecially made to the buyer’s order and were worthless, the seller wasallowed to deliver them and keep the purchase price.42 Having done so,

he was supposedly no longer enriched

Nevertheless, since courts have to bob and weave to come to the rightresult, it is not surprising that, in some of the cases to which Dawsonobjected, they came to the wrong one In one case, for example, the de-fendant employed the plaintiff to drill a hole on his land to the depth

of 800 metres The plaintiff could not even recover his costs, because, thecourt said, the hole was worthless, and therefore the defendant had notbeen unjustly enriched.43 Although the case is an old one, Dawson ar-gued that there is no logical escape from this result under German law.44There is not if we imagine that the defendant is liable solely because

he has been enriched We should say instead that even if the contract isvoid, those who order holes to be dug should bear the normal risks ofdoing so

Moreover, it is worth noting that American courts have sometimes madethe same mistake, not in asking whether enrichment has disappeared, but

in asking whether the defendant was ever enriched at all In one case,the defendant had contracted for the plaintiff to carve and erect a stonemonument The contract was unenforceable The plaintiff was allowed torecover in unjust enrichment for having dug a hole for the monumentbut not for the work he had done carving the stone, supposedly becausethe hole benefited the defendant but the carving did not since it hadnot been finished and the stone had not been erected.45 According tothe approach I am suggesting, he should have recovered for both, since

40 RGZ 94, 253, 255 41Lorenz: in Staudinger, § 831, n 46.

42 RGZ 118, 185, 188 Here, the machines had not yet been delivered, and so the court

did not apply the Saldotheorie; rather, it allowed the manufacturer to transfer the

machines to the buyer in satisfaction of his claim for unjust enrichment.

43 RG, 1911 JW 756. 44Dawson, ‘Erasable Enrichment’, 312–13.

45 Dowling v McKenny 124 Mass 478 (1878).